Github user morenn520 commented on the issue: https://github.com/apache/spark/pull/17238 @squito Actually it's a misconfiguration in our yarn. There are thousands of nodes in our production yarn. It will be quite frequent to rack/unrack nodes, so as to make something missing easily. It would be better if spark does something special to avoid our misconfiguration. In addition, I think unknown host should be recognized as different racks. But if MR/tez may do nothing special to this problem, it is reasonable for this problem not to be fix. I hope @tgravescs would double check whether tez/MR do anything special to DEFAULT_RACK.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org